A Theoretical and Experimental Study of the Transformational Process of Iconic and Verbal Information in Audiovisual Instructional Materials

本研究的目的在于发展一种视听教材的分析技术，包括符号信息和语言信息的处理。对53档日本和外国教育电视节目进行了不同类别的听觉和视觉成分分析。假设发送者想要传达的信息量与每个镜头的呈现时间长度成正比，因此以秒为单位测量的镜头长度由听觉和视觉呈现时间分量组成。引入复合指数模型，利用与听觉和视觉呈现时间分量相关的alpha和beta两个参数来近似视频节目镜头长度的各种分布。所有方案的理论分布与观测分布之间的一致性很好。为了探讨这两个参数的实际意义，定性和定量分析……在演讲者的数量、主题的内容、演讲的相对数量和“媒体复杂性”方面进行了更多的研究。这些结果都有利于建立视听模态的双分量指数模型。为了便于镜头分析技术和视频节目中图像的时间采样，还开发了一种能同时显示16幅分割图像的多画面显示系统。基于双分量指数模型，导出了以等时间间隔进行时间抽样的投篮百分比的理论估计公式，该公式在所有程序的期望值和实测值之间具有很好的一致性。对于大多数程序来说，超过95%的射击是由时间采样技术保证的，间隔时间小于5秒。讨论并尝试了这种多画面呈现技术在教学活动研究中的各种应用。少

The present study aimed at developing an analytical technique of audiovisual instructional materials in terms of iconic and verbal information processing. Fifty-three Japaneses and foreign educational TV programs were analyzed into auditory and visual components of different categories. The amount of information intended to convey by the sender was assumed to be proportional to the time length of presentation for each shot, so that a shot length measured in second is composed of auditory and visual presentation time components. A compound exponential model was introduced to approximate a variety of distributions of shot lengths of the video programs in terms of the two parameters alpha and beta respectively related to the auditory and visual presentation time components. The agreement between the therectical and the observed distributions for all the programs was very good. In order to explore the practical implications of the two parameters, qualitative as well as quantitative analysi … More s of shoes was carried out in terms of number of presenters, content of subject, relative amount of lecture presentation and "media complexities". All these findings were fairly conducive to the two component exponential model of audiovisual modalities. A multipicture display system which enables to present 16 partitioned pictures simultaneously was also developed to facilitate shot analytic technique and time sampling of pictures from video programs. Based on the two component exponential model, a theoretical formula for estimating the percentage of shots sampled by the time sampling with equal time intervals was derived and the formula found an excellent agreement between the expected and the observed values for all the programs. For most programs more than 95% of the shots were secured by the time sampling technique with less than 5 second intervals. Various applications of this multipicture presentation technique to the study of teaching-learning activities were discussed and attempted. Less